Senate Health Committee Issues Testimony From Harvard University
"Chairman Alexander, Ranking Member Murray, and Members of the
"My name is
The Nature of the Problem
"Administrative expenses are those expenses that are not directly associated with providing goods and services to people in need of care. There is no account kept on the amount of administrative expense of
"These estimates suggest that administrative expenses range from 15 to 30 percent of medical spending.
"1,2 To put this amount in perspective, even the smaller estimates suggest that administrative costs account for twice what
"Beyond the amount of money spent on administrative costs are the hassles associated with administration. The average
"The level of administrative expense in
"The bulk of administrative expenses are for 'billing and insurance related' (BIR) services. When people think of administrative expense, they often jump to activities in insurance companies. This is a part of the total, but only a part. Two-thirds of administrative expenses occur in offices of physicians, hospitals, and other care providers.7
"Administrative costs are a form of economic "arms race." Pushed by businesses and individuals to reduce spending, insurers introduce requirements providers must fulfill before they can get paid. These additional requirements cost the insurer money to enforce, but are worth it in the savings from not paying out additional claims. In response to new rules, providers hire additional personnel to maximize the amount they are reimbursed. Witnessing this, insurers beef up rules yet again, putting in place additional requirements for payment. The net effect is a spiral of cascading administrative costs on both side of the market, with no benefit to patients and no net benefit to insurers or providers.
"A depiction of the processes involved in BIR services in provider offices is shown in Figure 1, taken from Tseng et al.8 The activities include verifying a patient's eligibility for services; submitting bills in an appropriate format; reviewing those submissions; submitting documentation required for pre-authorization purposes; collecting copayment or coinsurance from patients; and providing quality information and other documentation about the outcome of the procedure. The typical hospital spends nearly
(Figure 1 omitted)
"Figure 2 shows the extent to which the activities in Figure 1 have been automated, using data from the
(Figure 2 omitted)
Steps to Reduce Administrative Expense
"The goal of policy is to reduce administrative costs, but to do so in a smart way. It is not that we want to eliminate the functions that administrative costs serve. Verifying that people are eligible to receive care, that reimbursement is accurate, and that fraud and abuse are prevented are important goals. Rather, the idea is to conduct these processes more efficiently.
"Administrative costs are not a monolithic, so there's not a single solution that will reduce them. However, there are number of actions that would materially reduce administrative costs.
Reducing Severity Adjustments
"A significant portion of administrative costs is associated with measuring the severity of a patient presenting for treatment. For example, a patient presenting to the emergency department for treatment will be coded into one of five different severity levels (99281-99285) based on the nature of the illness or injury of the patient and their past history. The underlying rationale for this differentiation is sound: it takes more resources for an emergency department to treat a more severely ill patient. However, the administrative requirement of billing in this system is extremely high. For example, a patient with a history of high blood pressure or diabetes will often move into a more severe category than one without those conditions. Thus, there are people whose job it is to search the records of every emergency department patient to look for whether every patient has a history of conditions which would bump the patient into a more lucrative reimbursement category.
"The emergency department example carries throughout medical care system. For a large share of medical care goods and services, the health care system creates enormous administrative cost by differentiating payments according to the severity of the patient's illness and background.
"The natural solution is to limit the extent of differentiation. For example, payers could have one code for emergency department admissions instead of five, and similarly for other medical care goods and services. CMS recently announced its intention to implement such a policy for evaluation and management visits, moving from five billing categories to just two.
"There are two potential drawbacks to reduced differentiation of payments with severity of illness. First, removing additional payment for more severely ill patients makes some patients with severe illness unprofitable. This may induce providers to discourage such patients from seeking care, for example by turning them away or making it difficult to schedule appointments. I suspect this concern is minor, and in any case steps can be taken to manage it. Recall that most providers are willing to care for even patients who bring no revenue (the uninsured); their mission justifies this activity. Thus, selection is less of a concern with providers than with insurers. And some carve outs to the no-severity adjustment rule can be created. For uncommon but expensive items - complex surgeries, for example - it makes sense to retain a severity adjustment; the administrative costs are low relative to the amount of money involved in creating winners and losers. Finally, it is possible that alternative risk adjustment models could be employed that address most of what the severity adjustment covers but without the detail of measuring the full set of past conditions. For example, patient age, gender, and zip code are routinely collected and are correlated with a host of risk factors. Even a simple medical factor such as whether a patient was hospitalized in the past year would provide significant risk adjustment without involving high collection burden.
"The second potential drawback is that severity-neutral payments will transfer resources from providers that see more complex patients to providers that see less complex patients. The key to addressing this concern is to ensure that enough of the savings from administrative simplification flow to providers, so that losses to such providers can be offset by enhanced revenues. Imagine that we reduce administrative costs in hospitals by half, or 5 percent of total hospital spending.
"Insurers could split the resulting savings with providers, for example cutting payments by only 2.5 percent. This additional surplus would almost certainly compensate the providers that lose money because their true patient mix is more severe than average.
"On balance, therefore, I believe that current severity adjustments are substantially inefficient relative to a simpler system without such detailed risk adjustment but with workarounds for some limited number of cases.
Standardizing Pre-Authorization Requirements
"A second reform that would reduce administrative costs is standardizing the documentation required for pre-authorization of services. A typical insurer will have a multitude of policies regarding what findings must be documented before it will authorize further treatment. For example, an MRI and physical therapy might be required before orthopedic surgery. Some such requirements are natural and beneficial, but there are far too many different requirements. It is not just that each insurer has their own pre-authorization requirements. Rather, each insurer has multiple different pre-authorization requirements, varying for each specific business they insure or public program they participate in. I once had a provider system show me the manual it keeps to bill radiology services alone; it was over a foot high.
"Complying with these requirements involves enormous expense. Armies of computer programmers and manual reviewers are employed by both insurers and providers to keep up with the changes. Further, the information required for the pre-authorization is often not easily accessible. The relevant information is in the physicians' electronic medical record, but there is no easy way for the electronic medical record to convey that information to the insurer's billing system. As a result, the process involves people. A person in the provider's office accesses the electronic or paper medical record, xeroxes the relevant pages, and faxes them over to the insurer. Different people in the insurance company then need to look at the information and document that the information satisfies the necessary requirements.
"Standardizing pre-authorization requirements would be a major step forward. One might imagine that insurers and providers could live with two options: a more generous policy for payers willing to spend more, and a more restrictive policy for payers on a tighter budget. Providers could then focus on a small number of metrics associated with demonstrating applicability of the services under these two regimes. Variation from the standard policies would not be prohibited but could be discouraged, perhaps by requiring the payer to pay for the additional administrative expense they impose for both insurers and providers by deviating from these rules.
Integrating Medical Record and Billing Systems
"There is another way to view the previous example about the difficulty of pre-authorization requirements, and that is the inability of some computer systems to talk to others. Part of the reason for people to be engaged with billing is because electronic medical records which record clinical information have no way to communicate information to payment systems run by insurers. Thus, when an insurer requires documentation of a particular diagnosis or prior treatment, it requires people to be involved. Normally, we think of computers as making up for the limitations of people. In health care, it is people who make up for the limitations of computers.
"By contrast to health care, consider what happens when a person shops at
"A related issue occurs with quality assessment required for many pay-for-performance systems. Almost all payers, including public programs, have some pay-for-performance incentives built into their contracts, for example additional money associated with meeting guideline care for people with chronic disease. Information on the quality metrics is often in the electronic medical record, but that is not the format it needs to be in for payment purposes. As a result, providers spend a good deal of time, effort, and money pulling information from electronic medical record systems and putting them in a format appropriate for pay-for-performance calculations.
"Technologically, there is no reason why electronic medical record systems cannot interface with billing systems or automatically submit information for quality assessment. However, there are few incentives for existing firms to make this happen. Providers do not wish to give insurers access electronic medical records, because they consider them proprietary. Each individual insurer has little incentive to invest in a system that is more conducive to provider systems, since doing so for a single practice involves large costs and little gain. Makers of electronic medical record systems have incentives to keep their systems exclusive, so that it is more difficult for providers to switch from one company to another. Thus, we are in a situation where costs remain high even though everyone recognizes that they could be reduced.
"The solution to the technological interoperability can be solved through either public or private actions. In the public sector, standards regarding health information technology could be modified so that select information flow from electronic medical record systems to billing systems is required. Most of the federal effort devoted to interoperability has focused on increasing access to clinical information by patients and providers. For example, everyone agrees that a person with a medical record at one organization who visits a provider at a second organization should be able to have their record read at the second provider. However, much less attention has been devoted to the links between medical records systems and billing systems.
"A private sector solution might involve something like the credit card industry, where intermediaries read information from electronic medical records and send the compiled information to insurers in the appropriate format. The intermediaries would take a common set of information from providers - the universe of information that is required - and then parcel out the information as required. As an analogy, consider the world of retail trade. One of the amazing features about retail is that the smallest stores can process the same payment methods as the largest stores. The reason for this is that firms such as
What Federal Policy Can Do
"Administrative costs have fallen in many industries throughout the economy. The retail sector was noted above. But credit cards are just the tip of the iceberg. Other examples include Universal Product Codes (UPCs) to make checkout less expensive, electronic sales for many goods, and employment of sophisticated information systems to reduce distribution and inventory costs. Another example is the financial services industry. Trillions of dollars are transmitted electronically each day, with barely any administrative cost. To a great extent, this is because the technology for doing so has been standardized.
"In each of these industries and others, there is a common theme to reducing administrative costs: administrative costs fall when there is a dominant player that forces standardization. In retail trade, standardization came about to a great extent because of the activities of
"There is only one organization in health care that is large as the
"What the federal government does not have is the mandate to do so.
"Both payment reform and IT promotion are important areas. My suggestion is not that the federal government not focus on these areas. Rather, I propose that each be coordinated with a third goal: creating and implementing a plan to reduce the administrative costs of medical care. To be as specific as possible, I propose that:
"
"
"Of course, one should not have blind faith in the ability of the federal government to coordinate in new areas. The disastrous opening of the Health Insurance Exchanges gives everyone pause about the wisdom of proposing federal action. On the other hand, the federal government has been a leader in many areas. Payment reform had no widescale implementation before recent federal actions, and the rollout of many payment models has gone well. And within the area of administrative simplification, Medicare was a leader in requiring claims to be submitted electronically. That explains a good part of why claims submission is almost fully electronic.
"The reality of the situation is this: unless the federal government leads the way,
* * *
Footnotes:
1 Yong PL, Saunders RS, Olsen L, eds. The healthcare imperative: lowering costs and improving outcomes -- workshop series summary.
2 Jiwani, Aliya,
3 Cutler, David M,
4 Casalino Lawrence P.,
5 Shanafelt, Tait D.,
6 Cutler, David M., and
7 Yong et al., op cit.
8 Tseng, Philip,
9 CAQH, 2017 CAQH Index: A Report of Healthcare Industry Adoption of Electronic Business Transactions and Cost Savings, 2018.
10 Yong et al., op cit.
11 Cutler, David M., "
12 Johnson,
Senate Health Committee Issues Testimony From Alaska State Hospital & Nursing Home Association
Refugio museum still feeling effects of Harvey
Advisor News
Annuity News
Health/Employee Benefits News
Life Insurance News